Disagreement in molecular weight determinations of dentin phosphoprotein

Abstract

The molecular weight of highly phosphorylated phosphoprotein front rat incisor dentin was estimated by three different methods: sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, gel chromatography in 6 M guanidinium chloride (GdnHCI) and analytical ultracentrifugation at high salt concentration. SDS-POlyacrylamide gel electrophoresis revealed an abnormal migration of the phosphoprotein and its enzymatically dephosphorylated derivative in comparison to standard proteins. Depending on the acrylamide concentration in the gel (7.5–20%), the apparent molecular weight of the phosphoprotein varied from 90000 to 27000. Similar results were obtained for the daphosphorylated protein. In gel chromatography using 6 M GdnHCI as eluent the phosphoprotein had an elution volume corresponding to that of a standard protein with a molecular weight of 67000, while the enzymatically dephosphorylated phosphoprotein showed an apparent molecular weight of 30000. The phosphoprotein behaved non-ideally in equilibrium sedimentation runs. The apparent molecular weight was strongly concentration-dependent and the extrapolation to zero concentration was uncertain. However, after enzymatic dephosphorylation the concentration dependence disappeared and a molecular weight of 28000 could be calculated. Since the phosphate groups represent 26% of the phosphoprotein by weight, the true molecular weight of the highly phosphorylated phosphoprotein components from rat incisor dentin should be 38000. The study shows that it is not possible directly to estimate the molecular weight of this type of protein by standard methods elaborated for normal globular proteins.